Agent Memory Guard
AI Agents & AssistantsRuntime defense layer that screens every read/write to an AI agent's memory, blocking prompt injection, secret leakage, and integrity tampering across sessions
Features
- Screens all reads and writes to protect against prompt‑injection attacks
- Detects and redacts or blocks secret/PII leakage from agent memory
- Provides SHA‑256 integrity baselines for protected keys (tamper detection)
- Enforces declarative YAML policies with actions such as allow, redact, quarantine, block
- Offers point‑in‑time snapshots and rollback for forensic recovery
Recent releases
View all 2 releases →Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Install & Platforms
Install via
pip
